A system for recognition and description of graphics
- 6 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 255-259
- https://doi.org/10.1109/icpr.1988.28217
Abstract
A system for automatic recognition and description of graphical primitives and their interconnections in a mixed text/graphics image is described. The input to the system is a high-resolution binary image obtained by scanning paper-based documents. The Hough transform is applied to the connected components of the image to locate and separate text strings of various font size and orientation. The graphics in the segmented image is processed to represent thin entities by their core lines and thick objects by their boundaries. Lines of various types are identified and segmented into straight line and curved line segments during this process. The line segments and their interconnections are analyzed to locate minimum redundancy loops which are adequate to generate a succinct description of the graphics. Various hatching and filling patterns in the image are identified and described. The algorithm has been implemented as software, and its robustness and performance have evaluated tested using a number of test images. The results for a simple test image are reported.Keywords
This publication has 4 references indexed in Scilit:
- Knowledge-Based Functional-Symbol Understanding In Electronic Circuit Diagram InterpretationPublished by SPIE-Intl Soc Optical Eng ,1986
- From Paper Drawings to Computer-Aided DesignIEEE Computer Graphics and Applications, 1985
- Block segmentation and text extraction in mixed text/image documentsComputer Graphics and Image Processing, 1982
- A Modular System for Interpreting Binary Pixel Representations of Line-Structured DataPublished by Springer Nature ,1982